Corsair Link reporting radical spike with CPU Core #1

Hey folks,

 

First time posting here, and i'm a bit stumped i'm afraid and could use some expertise.

 

I am finding that my Corsair Link running the H100i v2 is showing rapid, and radical spikes for my i7 6700k Core#1 temp.

 

At idle, the core temp is 30c, but then frequently just loading up the start of a program (for example Unigine benchmark), it'll immediately spike to 70-80c within seconds causing my LED's to turn red as they are set to do that at 80c for CPU temp in Link.

Immediately exiting the Unigine program for example, will decrease the CPU Core#1 temp back down within seconds to the 30c idle again. Surely, this can't be right?

 

My h100i v2 temp is maintaining around 44c with this immediate load, from an idle temp of 34c, so that seems reasonable.

 

Some additional notes:

1) My Asus Qfan BIOS has CPU fan set to 100% PWM manual mode, to ensure Link controls the fans.

2) H100i v2 fans (upgraded to ML120's) are linked through the Y splitter and set to CPU slot on mobo.

3) The ambient temperature in my house is 21c (71f)

 

Any thoughts would be greatly appreciated!

If it is the same core every time, chances are you are looking at a monitoring glitch rather than actual temp changes. Are you running anything else at the same time as Link? (HWMonitor, AI Suite, Real Temp, AIDA, etc.).

So previously I didn't have any additional monitoring programs, but I just installed HWMonitor to see if it synced up with what Link was saying, and now i'm a fair bit more concerned on my cpu temp.

 

I decided to boot up Path of Exile, and what you'll see from the image I've posted is that I went from ~30c CPU temps to over 60c almost immediately on boot of the game. Awhile longer, and it surpassed 80c. Shutting the game down, dropped it back to ~40c almost instantaneously too.

 

Do I potentially have an issue with the cooling system? This still doesn't quite seem 'right' to spike that quickly.

 

Additional note - I have not done any overclocking of any kind, and used the default thermal paste that was pre-applied on the h100iv2.

Just ran another test - I booted up Dangerous Golf, and played maybe 5 min, and my 80c CPU limit was tripped again. However in correction to my original post, it's pretty clear now that all of the CPU temps are rising.

 

This time I left both HWMonitor and Link running to screenshot right away. Attached is what those results were.

 

Now having run this PC for the last year and a half, I've never hit these temps before with much more aggressive games (I didn't closely monitor regrettably, but never had the 80c max limit happen). This is something brand new over the past 2 weeks, and originally I thought maybe it was a firmware update/software side, but this seems to indicate possibly not?

Oh, you've definitely got issues. Your coolant temperature shouldn't be at 46.6C either.

 

Question 1: What is your room/ambient temperature?

 

Question 2: Is your radiator configured as exhaust? If so, what other fans do you have going? I suspect that you have some pretty serious airflow issues.

Well thats no good to hear :(

 

My ambient temp hasnt fluctuated since my original post, so its at roughly 21c (72f) inside.

 

Ive got what should be a more than adequate airflow. In the Phanteks Enthoo case, I have 2x 140mm intake at front, 2x 120mm Corsair ML120 fans on the radiator set to exhaust (arrow on the fan pointing at radiator). And 1x Noctua 140mm fan on rear for exhaust.

Yes, this is a little different than 1 core value out of order and all your cores are too warm. Your coolant temp should never be in that 45C+ range unless you have a massive amount of case heat. From the last screen shot, it looks more like a 28C motherboard temp. If you see coolant temps +18C over your internal temp, something is wrong and it is not going to be fan speed.

 

Since you already verified the pump is set to 100% in BIOS and the Link shot indicates the pump is at speed as well, this is likely the beginning of some type of blockage in the cooler. Contact Corsair Tech support through the ticket system to get the RMA process started. There isn't much you can do. You take the pump end off and shake it, hoping to dislodge the blockage, but it is a sealed system and that debris will come around again at some point. Nevertheless, it might buy you a few more days. An easier thing to try might be tipping the case over on its side (while running) to accomplish the same thing. No guarantee either of these methods work, but you can try while waiting.

Ok, I will start a ticket. As this is my first liquid system, I am a bit curious as to what kind of blockage this would be? Does the coolant coagulate or something?

 

Is there any chance I just have a bad seating on the cpu with the cooler? I was thinking of pulling it and adding new silver and re-attaching. However I suppose that wouldnt explain the high coolant temp would it?

Sometimes a little bit of excess sealant mixed with anti-corrosives. No real way to know without opening it up (something you should not do). It's not a common occurrence, although this model seems to come up the most. It's nothing you did or could have prevented. Correct, the high coolant rules out the poor contact or bad TIM on the cold plate.